Revue d'économie et de statistique appliquée
Volume 22, Numéro 2, Pages 43-51
2025-12-09

Exploring The Dimensionality Of The Q-chat-10: A Latent Class Analysis Approach

Authors : Mohammed Zakariya .

Abstract

The objective of this study is to explore the latent structure of the Q-CHAT-10 by applying latent class analysis to a sample of toddlers residing in Saudi Arabia. Several model‐fit indices were used to identify the optimal number of classes, and entropy was employed to assess classification clarity (values ≥ 0.80 indicating acceptable separation). The results revealed the existence of three latent classes corresponding to low-risk, moderate-risk, and high-risk subgroups. Class 1, the low-risk group, comprises 31% of the sample and is characterized by low probabilities of autistic traits. Class 2, the largest subgroup (44%), represents the moderate-risk group and displays moderate probabilities of autistic traits across all items. Class 3, accounting for 25% of the sample, represents the high-risk group and is marked by consistently high probabilities of autistic traits. This study contributes to both health policy and clinical practice by identifying distinct autism risk subgroups among toddlers. Classifying children into low-, moderate-, and high-risk groups facilitates early, targeted interventions and more efficient allocation of healthcare resources, potentially reducing long-term burdens on families and health systems.

Keywords

Autism Spectrum Disorder (ASD) ; Q-CHAT-10 ; Latent Class Analysis ; Toddlers ; Saudi Arabia